BEHAVIOR Observations of Acoustic Signaling in Three Sharpshooters: Homalodisca vitripennis, Homalodisca liturata, and Graphocephala atropunctata (Hemiptera: Cicadellidae) 1 2,3 2 DIANA M. PERCY, ELIZABETH A. BOYD, AND MARK S. HODDLE Ann. Entomol. Soc. Am. 101(1): 253Ð259 (2008) ABSTRACT Observations and comparative data are presented on the acoustic signals of three sharpshooter (Hemiptera: Auchenorrhyncha: Cicadellidae: Cicadellinae) species native to North America. The acoustic signals of the glassy-winged sharpshooter, Homalodisca vitripennis (Germar), native to the southeastern United States, are compared with two smaller sharpshooters native to the western United States, Homalodisca liturata Ball and blue-green sharpshooter, Graphocephala at- ropunctata (Signoret). Each sharpshooter is a known vector of the plant pathogenic bacterium Xylella fastidiosa Wells et al. Male acoustic signals from all three species and female signals from H. liturata and G. atropunctata were recorded from host plant substrates. The H. vitripennis calls were recorded in the evening and morning, whereas H. liturata and G. atropunctata were recorded in the afternoon. Each species has a characteristic acoustic signal of which the male call structure is most complex in H. vitripennis and simplest in H. liturata. MaleÐfemale acoustic duets in H. liturata and G. atropunctata were recorded, and distinct differences were found in the reply latencies between male and female calls in these species. KEY WORDS acoustic signal, Homalodisca coagulata, leafhopper, sharpshooter, substrate vibration Sharpshooters are a group of xylophagous leafhoppers cadellini), are both indigenous to California. These in the tribes Proconiini and Cicadellini (Hemiptera: two native sharpshooter species also are efÞcient vec- Auchenorrhyncha: Cicadellidae: Cicadellinae). In the tors of various strains of X. fastidiosa (Severin 1949b, Americas, Ͼ1,400 species comprise these two tribes, Freitag and Frazier 1954, Hill and Purcell 1995, Krell with the greatest species diversity occurring in trop- et al. 2007), and they are considered pests of California ical and subtropical regions (Redak et al. 2004). Pro- agriculture. As a xylem feeder, H. vitripennis must coniine and Cicadelline leafhoppers range in size from consume large quantities of ßuid to meet its nutritional 3.4 to Ͼ20 mm, with the largest species in the Proco- requirements, and it is capable of consuming up to 100 niini (Redak et al. 2004). Homalodisca vitripennis times its body weight in a single day (Brodbeck et al. (Germar) (ϭHomalodisca coagulata [Say] [Takiya et 1993), thus producing large amounts of excreta and al. 2006])(Cicadellidae: Proconiini), the glassy- causing physiological damage to infested ßora winged sharpshooter, was accidentally introduced (Andersen et al. 2003, Hix 2004). Extremely high den- into California from its native range in southeastern sities of this insect in foliage result in copious amounts United States in the 1990s (Sorensen and Gill 1996), of excrement that can “rain” down on humans, other and it is a serious pest due to its extensive host plant plants, and parked vehicles. This “honeydew” excre- range and capacity to vector various strains of the ment dries leaving a crusty white-washed Þlm; thus, H. lethal plant bacterium Xylella fastidiosa Wells et al. vitripennis can be considered a substantial public nui- (Blua et al. 1999, Redak et al. 2004). Homalodisca sance in impacted areas (Grandgirard et al. 2006). H. liturata Ball (Cicadellidae: Proconiini; congener to H. vitripennis also has been unintentionally introduced vitripennis), and blue-green sharpshooter, Grapho- into a number of South PaciÞc islands; and in French cephala atropunctata (Signoret) (Cicadellidae: Ci- Polynesia (Society Islands), the extremely high den- sities of this insect required a classical biological con- 1 UBC Botanical Garden and Centre for Plant Research, University trol program, which was successfully implemented in of British Columbia, 6804 Southwest Marine Dr., Vancouver, BC, V6T 2004 by using an egg parasitoid, Gonatocerus ashmeadi 1Z4 Canada. 2 Department of Entomology, University of California, 3401 (Hymenoptera: Mymaridae) (Grandgirard et al. Watkins Dr., Riverside, CA 92521. 2006). 3 Current address: Division of Organisms and Environment, De- H. vitripennis, H. liturata, and G. atropunctata differ partment of Environmental Science, Policy, and Management, Uni- in ecology and size, with adults of each species mea- versity of California, Berkeley, Kearney Agricultural Center, 9240 S. Riverbend Ave., Parlier, CA 93648 (corresponding author, e-mail: suring Ϸ11Ð13 mm (Turner and Pollard 1959), 11 mm [email protected]). (Powers 1973), and 6Ð7 mm (DeLong and Severin 0013-8746/08/0253Ð0259$04.00/0 ᭧ 2008 Entomological Society of America 254 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol. 101, no. 1 1949), respectively. H. vitripennis can be found in both no prior insect exposure, and they were subjected to coastal and irrigated xeric habitats in California, and it only one sharpshooter species. Acoustic signals were is polyphagous (Redak et al. 2004). H. liturata occurs recorded as substrate vibrations by using optimized in both coastal and xeric habitats and also has an (Percy and Day 2005, Percy et al. 2006) substrate extensive list of host plants (Powers 1973). G. pickup methods developed by Claridge et al. (1985). atropunctata is found mainly in riparian vegetation The gramophone cartridge was positioned so the sty- consisting of favored oviposition plants, such as wild lus made contact with the surface of the primary stem grape, blackberry, stinging nettle, and mugwort (De- of the potted basil plants. Signals from the cartridge Long and Severin 1949). H. vitripennis and H. liturata were ampliÞed 10ϫ by using an ED1241 differential are bivoltine in southern California (Blua et al. 2001) ampliÞer (designed and constructed by C. Hardy, De- with adults living on average 2Ð3 mo (Turner and partment of Electronics and Electrical Engineering, Pollard 1959); G. atropunctata is univoltine in southern University of Glasgow, United Kingdom) and re- California with adults living up to 10Ð12 mo (Boyd and corded at a sampling rate of 44,100 samples per second Hoddle 2006). Each sharpshooter completes Þve in- (44.1 kHz) on digital audio tape (model TCD-D8, stars (Severin 1949a, Turner and Pollard 1959, Powers Sony DAT tape recorder). 1973). Acoustic signaling observations were conducted in Cicadellids spend their entire lives on plants, and an isolated and relatively quiet room maintained at they communicate by vibrational signals transmitted Ϸ25ЊC, with ßorescent lighting supplemented by am- via the host plant substrate (Claridge 1985, Bailey bient sunlight from nearby windows. One or more 2003, Cokl and Virant-Doberlet 2003, Percy and Day females were placed on the caged basil plant and 2005). Acoustic signaling in H. vitripennis has been allowed to settle. Additional females or males were recorded, but it is not described (UCANR 2005). then added as required. Because only one acoustic Acoustic signaling in H. liturata and G. atropunctata signal from a greenhouse-reared adult H. vitripennis had not been examined before this study. Conse- was recorded after Ϸ4 h of monitoring (with a total of quently, the objectives of this study were to record the 10 males and Þve females), Þeld-collected H. vitrip- acoustic signals of H. liturata and G. atropunctata and ennis adults from Citrus orchards were added to the compare those signals to similarly recorded signals for observation arena. These newly added H. vitripennis H. vitripennis. adults stimulated signaling among caged H. vitripennis adults on basil. H. liturata and G. atropunctata did not require the addition of Þeld-collected material be- Materials and Methods cause these two species began communicating rela- Sharpshooter colonies were housed in a tempera- tively quickly after set up on basil. The acoustic signals ture-controlled greenhouse at 26.7 Ϯ 0.6ЊC, 24.8 Ϯ were recorded during a total of 6Ð10 h of observations 13.4% RH, and a photoperiod of 16:8 (L:D) h in mul- and acoustic monitoring for each of the three species. tiple 75- by 75- by 75-cm cages at the University of Signals from H. liturata were recorded in a single, California, Riverside, CA. Greenhouse populations continuous monitoring period of Ϸ6 h (8.45Ð14.30), were supplemented with Þeld-collected insects each whereas H. vitripennis and G. atropunctata signals year. Each sharpshooter species was reared on a va- were recorded in two monitoring periods on separate riety of plants, including Eureka lemon [Citrus limon days: H. vitripennis day 1, Ϸ10 h (11.45Ð19.30) and day (L.) Burm.f. ÔEurekaÕ; Sapindales: Rutaceae], wild 2, Ϸ4.5 h (6.30Ð11.00); G. atropunctata day 1, 4 h grape (Vitis girdiana Munson; Rhamnales: Vitaceae), (13.30Ð17.30) and day 2, 2 h (13.00Ð15.00). We used and basil (Ocimum basilicum L.; Lamiales: Lami- between six and 10 adults per monitoring period (two aceae). Basil plants were selected for use in acoustic to three females and three to seven males; Table 1). studies because all three sharpshooter species readily General activity, feeding, and abdominal and wing feed, mate, and oviposit on basil in colony cages movements of individuals during sound production (E.A.B., unpublished data). For all acoustic observa- was observed to determine whether sounds were pro- tions, Þfth instars were removed from the colonies, duced by male or female sharpshooters. However, sorted according